Ultra-high field (UHF) MRI (e 7 Tesla) promise of detecting human disease within the human body continues to grow [4]. However, there are several issues experienced at higher field strengths that hinder its clinical feasibility. These specific UHF MRI challenges addressed by this work are: (1) achieving magnetic field homogeneity, (2) reducing the global/local specific absorption rate (SAR) in biological tissue that are not subject dependent and (3) addressing concerns regarding the unclear RF safety assurance of the multi-transmit experiment due to inappropriate electromagnetic models for the estimation of the SAR at UHF. Such challenges limit the intervention and detection of human disease for 7T. This work focuses on tackling this issue for the human head. The limitations drastically hinder clinicians from using UHF MRI for early detection and intervention of neurodegenerative diseases and symptoms of brain damage in patients. The main goal of this work is to provide solutions through hardware and software to the clinical use of UHF MRI and master the understanding of its limitations. The specific aims of this project are: Aim 1: Develop an Anatomically Detailed Human Head Phantom to Evaluate B1+, SAR, and B1+ Temperature Mapping through Simulation and Experimental Studies at 7T. An anatomically detailed human head phantom is designed to have eight (8) refillable tissue compartments that hold electromagnetically equivalent biological properties of the conductivity, permittivity, and T1 measurements at 7T. The model's computational simulations are produced to evaluate B1, SAR, and B1 temperature ++ mapping. Lastly, experimental studies of the model are compared to both the simulation and in-vivo B1 + mapping of the head phantom and the human head in which the phantom was originally modeled. Aim 2: Design, Develop and Evaluate a Nonlinear B1+ Shimming MRI Parallel Transmission (PTx) Optimization Algorithm to achieve the B1+ uniformity through Simulation and Experimental Studies at 7T. A novel method of RF excitation pulses are produced from a Parallel Transmission (PTx) tool to perform B1 shimming on various regions of the human body using various nonlinear algorithms (i.e. particle + swarm). The system is designed to reduce the required time to address the limitations though parallel computing techniques using software and hardware (i.e. GPUs). Aim 3: Evaluate the Performance of Multi-Row Transceiver Arrays using Parallel Transmission and/or B1+ Shimming for B1+ maps and SAR maps analysis through Simulation and Experimental Studies at 7T. In order that this is achieved, computational electromagnetic modeling for a variation of anatomically detailed human head models (contractions/expansions and tissue addition/subtractions) within an RF coil (8- ch and 16-ch) is performed. Novel RF excitation pulses obtained from B1 shimming in a region are used to provide acceptable local and average SAR across the entire volume of various human head models. To test its real application, we analyze and compare the B1 field and SAR maps associated with each of the RF array elements for all the head models. Lastly, we perform in-vivo studies of acceptable RF excitation pulses on subjects to validate Tx array. At the completion of this project, it is our expectation that the combination of results collected from aims 1, 2, and 3 will further the UHF MRI community's understanding of 7T imaging. Progress in this proposed work will allow 7T MR imaging to be more enhanced in its ability to detect human diseases by yielding a more uniform 7T MR image.